HighPerformance Digital Library Classification Systems:

1
High-Performance Digital Library Classification
Systems
From Information Retrieval to Knowledge Management
PI Hsinchun Chen, The University of Arizona
DLI-2 All-Projects Meeting
Cornell, October 18-19, 1999
2
Research Goals

• Automatic generation of large-scale
  classification systems (CL)

• Integration of system and human-generated
  classification systems

• High-performance simulation and visualization of
  Object Oriented Hierarchical Automatic Yellowpage
  (OOHAY)

3
Testbed

• Geoscience Georef and Petroleum Abstracts
  (800K) and Georef thesaurus (26K terms)

• Medicine CancerLit (1M) and
  UMLS (250K concepts)

• The Web Indexable pages (10M)
  and Yahoo directory (250K nodes)

4
Partners
5
The Field
Knowledge Management/Knowledge Networking
Definition
The Knowledge Networking (KN) initiative focuses
on the integration of knowledge from different
sources and domains across space and time... KN
research aims to move beyond connectivity to
achieve new levels of interactivity, increasing
the semantic bandwidth, knowledge bandwidth,
activity bandwidth, an cultural bandwidth among
people, organizations, and communities.
6
The Field
Knowledge Management Functionality
(Source GartnerGroup, 1998)
Concept Yellow Pages
Retrieved Knowledge

• Clustering categorization table of contents
• Semantic Networks index
• Dictionaries
• Thesauri
• Linguistic analysis
• Data extraction

• Collaborative filters
• Communities
• Trusted advisor
• Expert identification

Semantic
Value Recommendation
Collaboration
7
Automatic Generation of CL Foundation from
NSF/DARPA/NASA Digital Library Initiative-1
8
Automatic Generation of CL Foundation from
NSF/DARPA/NASA Digital Library Initiative-1
9
Automatic Generation of CL Foundation from
NSF/DARPA/NASA Digital Library Initiative-1
10
Automatic Generation of CL Foundation from
NSF/DARPA/NASA Digital Library Initiative-1
11
Automatic Generation of CL Foundation from
NSF/DARPA/NASA Digital Library Initiative-1
12
Automatic Generation of CL Foundation from
NSF/DARPA/NASA Digital Library Initiative-1
13
Automatic Generation of CL Foundation from
NSF/DARPA/NASA Digital Library Initiative-1
14
Automatic Generation of CL
15
Automatic Generation of CL (Continued)

• Entity Extraction and Co-reference based on TREC
  and MUG

• Text segmentation and summarization based on
  Textile and Wavelets

• Visualization techniques based on Fisheye,
  Fractal, and Spotlight

16
Integration of CL

• Lexicon-enhanced indexing (e.g., UMLS Specialist
  Lexicon)

• Ontology-enhanced query expansion (e.g.,
  WordNet, UMLS Metathesaurus)

• Ontology-enhanced semantic tagging (e.g., UMLS
  Semantic Nets)

• Spreading-activation based term suggestion
  (e.g., Hopfield net)

17
High-performance Simulation and Visualization

• Algorithmic optimization and parallelization on
  NCSA supercomputers (time machine)

• Advanced, interactive 2D/3D visualization via
  Java, VRML, and OpenGL

18
From YAHOO! To OOHAY?
Y
A
H
O
O
!
Object
Oriented
Hierarchical
Automatic
Yellowpage
?
19
OOHAY Visualizing the Web
20
OOHAY Visualizing the Web
21
For project information and free download
http//ai.bpa.arizona.edu
OOHAY CI Spider, Meta Spider, Med Spider
1. Enter Starting URLs and Key Phrases to be
searched
2. Search results from spiders are displayed
dynamically
22
For project information and free download
http//ai.bpa.arizona.edu
OOHAY CI Spider, Meta Spider, Med Spider
3. Noun Phrases are extracted from the web ages
and user can selected preferred phrases for
further summarization.
4. SOM is generated based on the phrases
selected. Steps 3 and 4 can be done in iterations
to refine the results.
23
Digital Library Research on New York Times, Cover
article, Sep 30, 1999
24
DL Special Issues and Activities

• Second Asia DL Workshop, November 8-9, 1999,
  Taipei, Taiwan

• JASIS, 2000, forthcoming (Chen)

Berkeley (Wilensky), UCSB (Hill/Smith), Maryland
(Greene/Shneiderman), Xerox PARC (Baldonado), IBM
(Liu), Texas AM (Shipman/Furuta), NASA (Kaplan)